Data Science & AI
Our focus is delivering data-driven tools and methods to support horizon scanning and intelligence sharing across our stakeholders, and underpin evidence synthesis research in public health.
The Data Science & Artificial Intelligence Programme, design and develop intelligent internal and external facing tools that search, identify, and extract data across various open data sources to help derive knowledge and insights. We also make and share tools that underpin, streamline, and automate aspects of evidence synthesis research, including systematic, scoping and mapping reviews.
Within this Programme sits our internal research and development lab comprising a mixed team of data scientists, software developers, academic researchers, and analysts, all working collaboratively across the Innovation Observatory.
Current live projects:
- OpenScan: Our core, dynamic, database architecture in the cloud, which uses natural language processing techniques to extract and contain health innovation data from a variety of data sources. OpenScan is the engine powering many of our core tools and systems.
- ScanMedicine: An evolving online search tool allowing users to query and draw insights from multiple resources, including global clinical trial registries and the FDA’s medical device databases, simultaneously. Currently in Beta. Try the tool here.
- ScanNovus (working title): An online search tool that aims to help make it easier for people to find research about the off-label and unlicensed use of medicines. Currently in development.
- ScanPalliative (working title): An online search tool helping users to discover meaningful intelligence relating to palliative and end-of-life care from trial and publication data. Currently in development.
- IO Toolkit: An internal toolkit containing a suite of tools to support research within the Innovation Observatory and across its programmes. Tools include (but aren’t limited to) specialised search filters, interactive visualisation tools, and AI-based screening classifiers.
